loaders/common.py

import tempfile
import time 
import os
from utils import compute_sha1_from_file
from langchain.schema import Document
import streamlit as st
from langchain.text_splitter import RecursiveCharacterTextSplitter
from stats import add_usage
import re

def clean_chat_text(text):
    """Clean chat export text to remove special characters and format consistently"""
    # Remove non-printable characters
    text = ''.join(char for char in text if char.isprintable())
    
    # Clean up WhatsApp-style timestamps and phone numbers
    text = re.sub(r'\[\d{1,2}/\d{1,2}/\d{2,4},\s+\d{1,2}:\d{1,2}:\d{1,2}\s+[AP]M\]', '', text)
    text = re.sub(r'‪\+\d{2,3}\s*\d{3,10}\s*\d{3,10}‬', '', text)
    
    # Remove joining messages
    text = re.sub(r'joined using this group\'s invite link', '', text)
    
    # Remove extra whitespace
    text = ' '.join(text.split())
    
    return text

def process_file(vector_store, file, loader_class, file_suffix, stats_db=None):
    documents = []
    file_name = file.name
    file_size = file.size
    if st.secrets.self_hosted == "false":
        if file_size > 1000000:
            st.error("File size is too large. Please upload a file smaller than 1MB or self host.")
            return
            
    dateshort = time.strftime("%Y%m%d")
    with tempfile.NamedTemporaryFile(delete=False, suffix=file_suffix) as tmp_file:
        tmp_file.write(file.getvalue())
        tmp_file.flush()
        loader = loader_class(tmp_file.name)
        documents = loader.load()
        file_sha1 = compute_sha1_from_file(tmp_file.name)
    os.remove(tmp_file.name)
    
    chunk_size = st.session_state['chunk_size']
    chunk_overlap = st.session_state['chunk_overlap']
    text_splitter = RecursiveCharacterTextSplitter.from_tiktoken_encoder(chunk_size=chunk_size, chunk_overlap=chunk_overlap)
    
    documents = text_splitter.split_documents(documents)
    
    # Clean the text content before creating metadata
    docs_with_metadata = [Document(page_content=clean_chat_text(doc.page_content),
                                 metadata={"file_sha1": file_sha1,
                                         "file_size": file_size,
                                         "file_name": file_name, 
                                         "chunk_size": chunk_size,
                                         "chunk_overlap": chunk_overlap,
                                         "date": dateshort,
                                         "user": st.session_state["username"]}) 
                         for doc in documents]
    
    try:
        # Add debug logging before vector store addition
        print(f"Attempting to add {len(docs_with_metadata)} documents")
        print(f"Sample cleaned content: {docs_with_metadata[0].page_content[:200] if docs_with_metadata else 'No documents'}")
        
        vector_store.add_documents(docs_with_metadata)
        
        if stats_db:
            add_usage(stats_db, "embedding", "file", metadata={"file_name": file_name,
                                                             "file_type": file_suffix, 
                                                             "chunk_size": chunk_size,
                                                             "chunk_overlap": chunk_overlap})
    except Exception as e:
        print(f"Error adding documents to vector store:")
        print(f"Exception: {str(e)}")
        print(f"Input details:")
        print(f"File name: {file_name}")
        print(f"File size: {file_size}")
        print(f"File SHA1: {file_sha1}")
        print(f"Number of documents: {len(docs_with_metadata)}")
        print(f"Chunk size: {chunk_size}")
        print(f"Chunk overlap: {chunk_overlap}")
        print(f"First document preview (truncated):")
        if docs_with_metadata:
            print(docs_with_metadata[0].page_content[:500])
            
        # Additional debug info for vector store
        print(f"Vector store type: {type(vector_store).__name__}")
        raise